A comprehensive QA pipeline utilizing multiple specialized agents to plan, generate, run, and review tests.
Install
mkdir -p .claude/skills/qa-aidome-co && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/16865" && unzip -o skill.zip -d .claude/skills/qa-aidome-co && rm skill.zipInstalls to .claude/skills/qa-aidome-co
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Use when running an agentic QA pass on the AIdome Endpoint Switchboard extension — test planning, bug hunting, edge-case analysis, or pre-release verification. Wraps the gem-team-inspired QA agent pipeline (research → plan → generate → run → fix) plus adversarial review.Key capabilities
- →Inventory source files, existing tests, and framework
- →Create phased test implementation plans
- →Critique test plans for assumptions and edge cases
- →Generate tests phase by phase
- →Run tests and report results
- →Fix compile and test failures
How it works
The skill orchestrates a pipeline of specialized agents to perform research, plan, generate, run, and fix tests for the AIdome Endpoint Switchboard extension, including adversarial review.
Inputs & outputs
When to use qa
- →Testing new adapters
- →Running pre-release QA
- →Investigating flaky tests
- →Conducting edge-case analysis
About this skill
Agentic QA Skill
A unified entry point for the QA pipeline. Use this skill when you need a systematic quality pass rather than a single test tweak.
When to invoke
- Adding or substantially changing an adapter, profile, or orchestration flow
- Before cutting a release (Phase 4 / 5 of the extension lifecycle)
- Investigating a flaky or regression-prone test
- Auditing agent files themselves for governance gaps
- Any time a reviewer asks "did we cover the edge cases?"
Agent roster (see .github/references/qa-workflow.md for attribution)
| Agent | Role |
|---|---|
qa | Adversarial QA specialist — test plan + bug hunt |
critic | Challenges assumptions, finds logic gaps |
critical-thinking | Pre-implementation questioning (no edits) |
test-researcher | Analyze codebase for testability |
test-planner | Phased test implementation plan |
test-generator | RPI pipeline driver for generating tests |
test-fixer | Fix compile / test failures |
test-runner | Run tests, report results |
agent-governance-reviewer | Reviews agent configs for governance gaps |
test-engineer | Repo-native Vitest specialist (integration point) |
Recommended pipeline
1. test-researcher → inventory source files, existing tests, framework (Vitest)
2. test-planner → phased plan covering adapters, profiles, orchestrators
3. critic → critique plan: assumptions, edge cases, over-engineering
4. test-generator → generate tests phase by phase (delegates to test-engineer
for repo-specific Vitest + vscode mock patterns)
5. test-runner → npm test; report pass/fail
6. test-fixer → fix compile / assertion errors; loop ≤ 3 times
7. qa → exploratory / adversarial pass on merged result
8. agent-governance-reviewer → if any .agent.md file was touched
Local invariants the QA pipeline must enforce
npm run lint— zero errors; noconsole.loginsrc/npm run compile— zero TypeScript errors (strict mode)npm test— all unit tests + pre-release validation tests pass- All credentials only in
vscode.SecretStorage - Every adapter
configure()creates a timestamped backup before writing - URL scheme allowlist enforced; profile names sanitized
- Redaction utility applied before any log of sensitive values
If the QA pipeline cannot satisfy any of the above, escalate to the orchestrator; do not weaken the tests or skip them.
Outputs
- Test plan in the PR description or a scratch file
- New/updated
test/**files followingtesting.instructions.md - Findings report with Severity (Critical / High / Medium / Low)
- Go/no-go recommendation for release (Phase 5 gate)
See also
.github/references/qa-workflow.md— gem-team alignment + attribution.github/instructions/testing.instructions.md— Vitest conventions.github/agents/orchestrator.agent.md— orchestration contract
When not to use it
- →When not working on the AIdome Endpoint Switchboard extension
- →When a systematic quality pass is not required
Limitations
- →Specific to the AIdome Endpoint Switchboard extension
- →Relies on Vitest for testing
- →Requires `npm run lint` and `npm run compile` to pass
How it compares
This skill uses a multi-agent pipeline for systematic QA, including adversarial review and automated test generation/fixing, providing a more thorough and automated approach than manual testing.
Compared to similar skills
qa side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| qa (this skill) | 0 | 3mo | No flags | Advanced |
| customaize-agent:create-hook | 0 | 5mo | Review | Intermediate |
| dependency-upgrade | 26 | 5mo | Review | Intermediate |
| playwright-browser-automation | 29 | 8mo | Review | Intermediate |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
customaize-agent:create-hook
LAI-YEN-CHUN
Create and configure git hooks with intelligent project analysis, suggestions, and automated testing
dependency-upgrade
wshobson
Manage major dependency version upgrades with compatibility analysis, staged rollout, and comprehensive testing. Use when upgrading framework versions, updating major dependencies, or managing breaking changes in libraries.
playwright-browser-automation
lackeyjb
Complete browser automation with Playwright. Auto-detects dev servers, writes clean test scripts to /tmp. Test pages, fill forms, take screenshots, check responsive design, validate UX, test login flows, check links, automate any browser task. Use when user wants to test websites, automate browser interactions, validate web functionality, or perform any browser-based testing.
codex-skill
feiskyer
Use when user asks to leverage codex, gpt-5, or gpt-5.1 to implement something (usually implement a plan or feature designed by Claude). Provides non-interactive automation mode for hands-off task execution without approval prompts.
angular-best-practices
sickn33
Angular performance optimization and best practices guide. Use when writing, reviewing, or refactoring Angular code for optimal performance, bundle size, and rendering efficiency.
browser-tools
Whamp
Lightweight Chrome automation toolkit with shared configuration, JSON-first output, and six focused scripts for starting, navigating, inspecting, capturing, evaluating, and cleaning up browser sessions.